Comparative Analysis of SVM and ANN Classifiers using Multilevel Fusion of Multi-Sensor Data in Urban Land Classification

Vohra, Rubeena; Tiwari, K. C.

doi:10.1007/s11220-020-00280-9

Cited by 10 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data fusion is an effective way of synergistically combining information from various sources to better understand a given scene [13]. The fusion of MS and SAR data can sharpen the details of low-resolution data while providing complementary data observed from the same site [14]. In other words, the fusion of GF-3 and Sentinel-2 can not only help to generate images with rich spatial and spectral information but also contribute to better understand and explain the image area.…”

Section: Introductionmentioning

confidence: 99%

A Novel Image Fusion Method of Multi-Spectral and SAR Images for Land Cover Classification

et al. 2020

View full text Add to dashboard Cite

The fusion of multi-spectral and synthetic aperture radar (SAR) images could retain the advantages of each data, hence benefiting accurate land cover classification. However, some current image fusion methods face the challenge of producing unexpected noise. To overcome the aforementioned problem, this paper proposes a novel fusion method based on weighted median filter and Gram–Schmidt transform. In the proposed method, Sentinel-2A images and GF-3 images are respectively subjected to different preprocessing processes. Since weighted median filter does not strongly blur edges while filtering, it is applied to Sentinel-2A images for reducing noise. The processed Sentinel images are then transformed by Gram–Schmidt with GF-3 images. Two popular methods, principal component analysis method and traditional Gram–Schmidt transform, are used as the comparison methods in the experiment. In addition, random forest, a powerful ensemble model, is adopted as the land cover classifier due to its fast training speed and excellent classification performance. The overall accuracy, Kappa coefficient and classification map of the random forest are used as the evaluation criteria of the fusion method. Experiments conducted on five datasets demonstrate the superiority of the proposed method in both objective metrics and visual impressions. The experimental results indicate that the proposed method can improve the overall accuracy by up to 5% compared to using the original Sentinel-2A and has the potential to improve the satellite-based land cover classification accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

A Novel Image Fusion Method of Multi-Spectral and SAR Images for Land Cover Classification

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Multispectral remote sensing data is distinguished by its utilisation of narrow spectral bands that possess a comparatively wider bandwidth, facilitates the analysis of spatial attributes pertaining to ground substances (Vohra and Tiwari, 2020).…”

Section: Introductionmentioning

confidence: 99%

“…These limits are a result of similarities in spectral characteristics between various objects or the close proximity of objects in space. Therefore, in order to improve the accuracy of data evaluation, it is necessary to correctly analyse the properties of objects, including their con guration and spatial interconnections, along with their spectral response (Vohra and Tiwari, 2020).…”

Section: Introductionmentioning

confidence: 99%

Beyond Traditional Methods: Innovative Integration of LISS IV and Sentinel 2A Imagery for Unparalleled Insight into Himalayan Ibex Habitat Suitability

Dutta,

Sharma,

Joshi

et al. 2023

Preprint

View full text Add to dashboard Cite

Despite the progress made in remote sensing technology, the application of satellite imagery is predominantly limited to the field of conservation study. The utilisation of multispectral data from diverse sensors holds significant promise in the field of landscape mapping. However, it is imperative to consider the varied spectral and spatial resolution capabilities in order to achieve precise classification of wildlife habitats. The objective of our study was to provide a methodology for accurately classifying habitat types for the Himalayan Ibex (Capra sibirica) by utilising various satellite data. In order to tackle the issues related to both spectral and spatial aspects, we employed LISS IV and Sentinel 2A data. We then proceeded by integrating the LISS IV data with the Sentinel 2A data, taking into account their respective geometric information. By utilising a variety of supervised classification techniques, it was shown that the Random Forest (RF) approach had superior performance compared to the other algorithms. The classified image obtained by the integration of LISS IV and Sentinel 2A sensors demonstrated the highest level of accuracy, with an overall accuracy of 86.17% and a Kappa coefficient of 0.84. In order to delineate the suitable habitat for the Himalayan Ibex, we employed ensemble modelling techniques that incorporated Land Cover Land Use (LCLU) data from three distinct image types (namely LISS IV, Sentinel 2A, and Integrated image). Additionally, we incorporated other predictors including topographical features, vegetation types, soil and water radiometric indices. The integrated image demonstrated superior accuracy in predicting the suitable habitat for the Himalayan Ibex, compared to the other two LULC classes that were obtained from the other two mentioned images. The identification of suitable habitats was found to be contingent upon the consideration of two key factors: the Soil Adjusted Vegetation Index and elevation. The consequences of these findings are significant for the advancement of conservation measures, as the utilisation of precise classification methods facilitates the recognition of crucial landscape components. This pilot study offers a novel and important approach to conservation planning by accurately categorising LULC and identifying critical habitats for the Ibex. The utilisation of this technology significantly improves our capacity to conserve and safeguard the natural environment inhabited by many wildlife species within the mountainous ecosystem like the Himalayas.

show abstract

“…The data acquired through multispectral remote sensing is characterised by narrow spectral bands that possess a relatively larger bandwidth. Consequently, the gathered data can be employed to examine the spatial characteristics of ground substances (Vohra and Tiwari 2020). The limitations of a single source of satellite data in accurately extracting ground objects are attributed to spectral resemblance among different objects or spatial proximity between the objects.…”

Section: Introductionmentioning

confidence: 99%

“…The limitations of a single source of satellite data in accurately extracting ground objects are attributed to spectral resemblance among different objects or spatial proximity between the objects. Consequently, for the enhancement of data evaluation precision, it is imperative to appropriately construe object characteristics such as configuration and spatial interconnections, in conjunction with the spectral response (Vohra and Tiwari 2020).…”

Section: Introductionmentioning

confidence: 99%

Beyond Traditional Methods: Innovative Integration of LISS IV and Sentinel 2A Imagery for Unparalleled Insight into Himalayan Ibex Habitat Suitability

Dutta

Joshi

Kumar

et al. 2023

Preprint

View full text Add to dashboard Cite

Despite advancements in remote sensing, satellite imagery is underutilized in conservation research. Multispectral data from various sensors have great potential for mapping landscapes, but distinct spectral and spatial resolution capabilities are crucial for accurately classifying wildlife habitats. Our study aimed to develop a technique for precisely discerning habitat categories for the Himalayan Ibex (Capra sibirica) using different satellite imagery. To address both spectral and spatial challenges, we utilized LISS IV and Sentinel 2A data and integrated the LISS IV data with Sentinel 2A data along with their corresponding geometric information. Employing multiple supervised classification algorithms, we found the Random Forest (RF) algorithm to outperform others. The integrated (LISS IV-Sentinel 2A) classified image achieved the highest accuracy, with an overall accuracy of 86.17% and a Kappa coefficient of 0.84. To map the suitable habitat of the Ibex, we conducted ensemble modeling using the Land Cover Land Use (LCLU) of all three image types (LISS IV, Sentinel 2A, Integrated) and other predictors such as topographical, soil type, vegetation, and water radiometric indices. The integrated model provided the most accurate prediction of suitable habitat for the Ibex, surpassing the other two LCLU classes derived from individual images. The Soil Adjusted Vegetation Index (SAVI) and elevation were identified as crucial factors in identifying suitable habitats. These findings hold valuable implications for the development of effective conservation strategies, as accurate classification schemes enable the identification of vital landscape elements. By precisely classifying LULC satellite images and identifying crucial habitats for the Ibex, this pilot study provides a new and valuable strategy for conservation planning. It enhances our ability to preserve and protect the habitat of wildlife species in the mountain ecosystem of the Himalayas.

show abstract

Comparative Analysis of SVM and ANN Classifiers using Multilevel Fusion of Multi-Sensor Data in Urban Land Classification

Cited by 10 publications

References 29 publications

A Novel Image Fusion Method of Multi-Spectral and SAR Images for Land Cover Classification

A Novel Image Fusion Method of Multi-Spectral and SAR Images for Land Cover Classification

Beyond Traditional Methods: Innovative Integration of LISS IV and Sentinel 2A Imagery for Unparalleled Insight into Himalayan Ibex Habitat Suitability

Beyond Traditional Methods: Innovative Integration of LISS IV and Sentinel 2A Imagery for Unparalleled Insight into Himalayan Ibex Habitat Suitability

Contact Info

Product

Resources

About